Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.programming > #2358
| From | "BartC" <bc@freeuk.com> |
|---|---|
| Newsgroups | comp.programming |
| Subject | Re: Buddy System Memory Allocator |
| Date | 2012-10-19 15:04 +0100 |
| Organization | A noiseless patient Spider |
| Message-ID | <k5ro9b$cg2$1@dont-email.me> (permalink) |
| References | <_Iges.3$jY.2@fx25.am4> <1731845837372288843.910496pjb-informatimago.com@news.individual.net> |
"Pascal J. Bourguignon" <pjb@informatimago.com> wrote in message news:1731845837372288843.910496pjb-informatimago.com@news.individual.net... > "Mark" <mark@dibsco.co.uk> wrote: >> of memory and want a relatively small minimum block size. A bitmap of 32 >> bits will only give me coverage for 8k of memory if I use a minimum block >> size of 256 bytes. I am trying to figure out how to avoid having a bitmap >> that is hundreds or thousands of bits long. I'm sure that there must be a >> better way. > The simpliest is to keep the size with the allocated blocks: allocate a > word more than requested, to store the size, and return the addess of the > first byte after the size. You may want to still return aligned pointers > so you may want to store the size on a smaller alignment. He's complaining about having to use a 4 byte bitmap for every 8KB (1 bit per 256 bytes). Your way might need 4 bytes for every 256 bytes! And could introduce alignment issues (260 bytes per block also sound as a bit awkward). -- bartc
Back to comp.programming | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
Buddy System Memory Allocator "Mark" <mark@dibsco.co.uk> - 2012-10-13 17:32 +0100
Re: Buddy System Memory Allocator "BartC" <bc@freeuk.com> - 2012-10-14 16:33 +0100
Re: Buddy System Memory Allocator Pascal J. Bourguignon <pjb@informatimago.com> - 2012-10-19 12:24 +0000
Re: Buddy System Memory Allocator "BartC" <bc@freeuk.com> - 2012-10-19 15:04 +0100
Re: Buddy System Memory Allocator Pascal J. Bourguignon <pjb@informatimago.com> - 2012-10-19 16:37 +0000
csiph-web